Intelligent Comfort and Energy-saving Fan Coil Control System

Abstract

The present invention relates to the field of fan coil units, and particularly to an intelligent, comfortable and energy-saving fan coil unit control system. The intelligent, comfortable and energy-saving fan coil unit control system includes a target temperature prediction module, a two-way valve control module and a wind speed adjustment module. By adopting a target temperature prediction model, the present invention predicts the most suitable indoor temperature based on environmental data such as indoor and outdoor temperature and humidity. This prediction not only considers environmental factors, but also combines the user's historical preferences and behavior patterns, thereby achieving more accurate and personalized temperature control. At the same time, through the optimization of the PID control algorithm, the on-off time of the two-way valve can be dynamically adjusted according to the difference between the real-time monitored indoor temperature and the predicted target temperature, thereby reducing unnecessary energy consumption. This dynamic adjustment mechanism greatly improves the energy efficiency ratio of the air-conditioning system, achieving the reduction of energy consumption while ensuring indoor comfort.

Description

Technical Field

[0001] The present invention relates to the field of fan coil units, and particularly to an intelligent, comfortable and energy-saving fan coil control system. Background Art

[0002] In the field of building energy conservation, as the main energy consumption source, the efficiency and control method of the air conditioning system directly affect energy conservation and the comfort of the indoor environment. Most traditional fan coil systems adopt simple on-off control, lacking real-time response to changes in the indoor and outdoor environments, resulting in energy waste and indoor temperature fluctuations, and unable to meet the dual requirements of modern buildings for energy conservation and comfort.

[0003] With the development of intelligent technologies, more and more research has begun to focus on how to optimize the performance of the air conditioning system through advanced control strategies. The intelligent, comfortable and energy-saving fan coil control system proposed by the present invention aims to achieve more accurate and energy-saving temperature control and provide personalized indoor environment adjustment by integrating technologies such as target temperature prediction, PID control algorithm optimization, two-way valve intelligent control, and dynamic air volume adjustment. Summary of the Invention

[0004] The present invention uses a target temperature prediction model to predict the most suitable indoor temperature based on environmental data such as indoor and outdoor temperatures and humidity. This prediction not only considers environmental factors but also combines the user's historical preferences and behavior patterns, thereby achieving more accurate and personalized temperature control. At the same time, through the optimization of the PID control algorithm, the on-off time of the two-way valve can be dynamically adjusted according to the difference between the real-time monitored indoor temperature and the predicted target temperature, thereby reducing unnecessary energy consumption. This dynamic adjustment mechanism greatly improves the energy efficiency ratio of the air conditioning system and realizes energy consumption reduction while ensuring indoor comfort.

[0005] The intelligent, comfortable and energy-saving fan coil control system includes:

[0006] A target temperature prediction module, which is used to input a target temperature prediction set composed of indoor temperature data, outdoor temperature data, indoor humidity data, and outdoor humidity data into the target temperature prediction model, and output a predicted target temperature T, where the predicted target temperature T is the air conditioning set temperature within the currently preset control period;

[0007] A two-way valve control module, which is used to calculate and obtain an output P using the PID control algorithm according to the difference between the current indoor temperature data and the predicted target temperature; convert the output P into a duty cycle of the two-way valve on-off time; within the currently preset control period, apply the duty cycle of the two-way valve on-off time to control the opening and closing time of the two-way valve;

[0008] The wind speed adjustment module is used to adjust the wind speed according to the indoor temperature data and the operating state of the two-way valve. When the two-way valve is in the open state, compare the indoor temperature data at the currently obtained second monitoring time point with the predicted target temperature T. If the indoor temperature data is greater than T + 0.5, adjust the wind speed to high wind speed; if the indoor temperature data is greater than T and less than T + 0.5, adjust the wind speed to medium wind speed; if the current indoor temperature is less than T, adjust the wind speed to low wind speed. When the two-way valve is in the closed state, adjust the wind speed to low wind speed.

[0009] It also includes an intelligent comfort and energy-saving fan coil control method applied to an intelligent comfort and energy-saving fan coil control system. The specific steps are as follows:

[0010] S1: Obtain the indoor temperature data, outdoor temperature data, indoor humidity data, and outdoor humidity data at the current first monitoring time point, and form a target temperature prediction set with the obtained indoor temperature data, outdoor temperature data, indoor humidity data, and outdoor humidity data. Input the target temperature prediction set into the target temperature prediction model to output the predicted target temperature T, and the predicted target temperature T is the air conditioner set temperature within the currently preset control period.

[0011] S2: Based on the indoor temperature data obtained at the current first monitoring time point, calculate the difference between the current indoor temperature data and the predicted target temperature, and use the PID control algorithm to calculate and obtain the output P. Convert the output P into the duty cycle of the two-way valve switching time. Within the currently preset control period, apply the duty cycle of the two-way valve switching time to control the opening and closing time of the two-way valve.

[0012] S3: Obtain the indoor temperature data at the current second monitoring time point, and determine whether the two-way valve is in the open state at the current time. If the two-way valve is in the open state, compare the indoor temperature data at the currently obtained second monitoring time point with the predicted target temperature T. If the indoor temperature data is greater than T + 0.5, adjust the wind speed to high wind speed; if the indoor temperature data is greater than T and less than T + 0.5, adjust the wind speed to medium wind speed; if the current indoor temperature is less than T, adjust the wind speed to low wind speed.

[0013] If the two-way valve is in the closed state, adjust the wind speed to low wind speed.

[0014] Preferably, the target temperature prediction model is established based on a linear regression model, and the target temperature prediction model uses a formula to predict the value y of the predicted target temperature; x1, x2, x3, and x4 are all input features, namely indoor temperature data, outdoor temperature data, indoor humidity data, and outdoor humidity data; β0 is the intercept term; β1, β2, β3, and β4 are the weights of each input feature respectively.

[0015] Preferably, for the training of the target temperature prediction model, the specific operation steps are as follows:

[0016] Obtain a number of model training samples. Each model training sample includes indoor temperature data, outdoor temperature data, indoor humidity data, outdoor humidity data, and the air conditioner set temperature at the corresponding moment. Divide all model training samples into a model training set and a model test set. Use the model training set to train the target temperature prediction model with the goal of minimization, where n is the number of the model training set; i = 1, 2,..., n. Use the model test set to evaluate the target temperature prediction model and obtain the evaluation result. Set the training conditions. If the evaluation result meets the training conditions, output the trained target temperature prediction model; otherwise, continue to train using the model training set.

[0017] Preferably, for the setting of the control period, the specific operation steps are as follows:

[0018] Set the update period. At the beginning of each update period, record the time required for the indoor temperature to rise from T - 1 to T + 1 under the state of the two-way valve closed and low wind speed, and use the obtained time as the upper limit duration of the control period. With the goal of maximizing the temperature control accuracy and minimizing the on-off frequency of the two-way valve, use the swarm optimization algorithm to search for the optimal control period within the range of 0 to the upper limit duration, and use the obtained optimal control period as the control period of the current update period.

[0019] Preferably, in the operation of setting the control period, the swarm optimization algorithm used is the particle swarm optimization algorithm. The specific operation steps are as follows:

[0020] A1: Randomly generate a particle swarm within the range of 0 to the upper limit duration. The position of each particle represents a candidate control period.

[0021] A2: For any particle, initialize the velocity of the particle and use the current position as the individual best position of the particle. Calculate the fitness value F of the particle using the formula, where D represents the difference between the maximum temperature and the minimum temperature within the current candidate period, and L represents the duration of the current candidate period; and are the weights of and L respectively.

[0022] A3: Traverse all particles and use the position corresponding to the particle with the highest fitness value F as the global best position.

[0023] A4: For any particle, update the velocity and position of the particle and recalculate the fitness value. If the updated fitness value is higher than the fitness value of the individual best position of the particle, update the individual best position of the particle to the current position.

[0024] A5: Traverse all particles. If the fitness value of the individual best position of any particle is higher than the fitness value of the global best position, update the global best position to the individual best position of this particle;

[0025] A6: Repeat steps A4 - A5 until the maximum number of iterations is reached. The obtained global best position is the optimal control period.

[0026] Preferably, the specific operation of calculating and obtaining the output P using the PID control algorithm in step S2:

[0027] Calculate the difference between the current indoor temperature data and the predicted target temperature, and calculate and obtain the PID output P using the formula; where K

[0029] , d , p , min , i , i , max , d , p is the proportional gain; T i is the integral time constant; T d is the differential time constant; m represents that within the integral time constant, there are m control cycles, j = 1, 2,..., m; represents the difference between the indoor temperature data obtained in the j - th control cycle and the predicted target temperature in the most recent time period with a duration of the integral time constant; is the difference between the indoor temperature data and the predicted target temperature obtained in the previous control cycle; is the duration of the control cycle.

[0028] Preferably, convert the output P to the duty cycle of the two - way valve opening time; the specific operation of applying the duty cycle of the two - way valve opening time within the currently preset control cycle:

[0029] Obtain the range [P min , P max of the PID output P, and calculate and obtain the duty cycle N of the two - way valve opening time using the formula; within the current control cycle, the two - way valve is opened for a time and then closed for a time.

[0030] Preferably, the values of the proportional gain K p , the integral time constant T i and the differential time constant T d are all obtained through the particle swarm optimization algorithm.

[0031] The present invention has the following advantages:

[0032] 1. By adopting a target temperature prediction model, the present invention predicts the most suitable indoor temperature based on environmental data such as indoor and outdoor temperature and humidity. This prediction not only considers environmental factors but also combines the user's historical preferences and behavior patterns, thereby achieving more accurate and personalized temperature control. At the same time, through the optimization of the PID control algorithm, the on-off time of the two-way valve can be dynamically adjusted according to the difference between the real-time monitored indoor temperature and the predicted target temperature, thus reducing unnecessary energy consumption. This dynamic adjustment mechanism greatly improves the energy efficiency ratio of the air-conditioning system and realizes the reduction of energy consumption while ensuring indoor comfort.

[0033] 2. The present invention can automatically adjust the air-conditioning set temperature according to the user's personalized needs and the actual conditions of the indoor and outdoor environment, provide a customized indoor environment, and also improve the environmental quality by reducing temperature fluctuations. At the same time, the wind speed is adjusted according to the indoor temperature and the air-conditioning set temperature to reduce unnecessary energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic structural diagram of an intelligent comfort energy-saving fan coil control system adopted in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] In order to enable those skilled in the art of the present technology to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention.

[0036] Embodiment, an intelligent comfort energy-saving fan coil control system, as Figure 1 shown, includes:

[0037] A target temperature prediction module, configured to input a target temperature prediction set composed of indoor temperature data, outdoor temperature data, indoor humidity data, and outdoor humidity data into the target temperature prediction model, and output a predicted target temperature T, where the predicted target temperature T is the air-conditioning set temperature within the current preset control period;

[0038] A two-way valve control module, configured to calculate and obtain an output P by using the PID control algorithm according to the difference between the current indoor temperature data and the predicted target temperature; convert the output P into a duty cycle of the two-way valve on-off time; within the current preset control period, apply the duty cycle of the two-way valve on-off time to control the opening and closing time of the two-way valve;

[0039] A wind speed adjustment module, which is used to adjust the wind speed according to the indoor temperature data and the operating state of the two-way valve; when the two-way valve is in the open state, compare the indoor temperature data at the current obtained second monitoring time point with the predicted target temperature T. If the indoor temperature data is greater than T + 0.5, adjust the wind speed to high wind speed; if the indoor temperature data is greater than T and less than T + 0.5, adjust the wind speed to medium wind speed; if the current indoor temperature is less than T, adjust the wind speed to low wind speed; when the two-way valve is in the closed state, adjust the wind speed to low wind speed.

[0040] It also includes an intelligent comfort and energy-saving fan coil control method applied to an intelligent comfort and energy-saving fan coil control system. The specific steps are as follows:

[0041] S1: Obtain the indoor temperature data, outdoor temperature data, indoor humidity data, and outdoor humidity data at the current first monitoring time point. The indoor temperature data can most directly reflect the indoor thermal environment state, and the change of outdoor temperature data will affect the indoor temperature; indoor humidity will have a certain impact on the human body's perceived temperature. For example, when the indoor humidity is too high, it will inhibit the body's natural cooling mechanism, and at the same time make people feel the air is dull and intensify the sense of heat. Therefore, a lower air conditioner set temperature is required, and outdoor humidity will also affect people's physical feelings; form the obtained indoor temperature data, outdoor temperature data, indoor humidity data, and outdoor humidity data into a target temperature prediction set; input the target temperature prediction set into the target temperature prediction model to output the predicted target temperature T, which is considered the most suitable indoor temperature under the current environmental conditions. The predicted target temperature T is the air conditioner set temperature within the current preset control period. The control period is the time interval for adjusting the air conditioner set temperature, and the time between every two first monitoring time points is the control period;

[0042] S2: Based on the indoor temperature data obtained at the current first monitoring time point, calculate the difference between the current indoor temperature data and the predicted target temperature. This difference is the basis for the control system to make adjustments and reflects the deviation between the current indoor temperature and the air conditioner set temperature. Use the PID control algorithm to calculate and obtain the output P. The output P is the calculation result of integrating the proportional, integral, and differential parts; convert the output P into the duty cycle of the two-way valve switching time; within the preset control period, the system controls the opening and closing time of the two-way valve according to the calculated duty cycle; if the duty cycle is high, the two-way valve will be open for a longer time to quickly adjust the indoor temperature; if the duty cycle is low, the opening time of the two-way valve will be shortened to keep the indoor temperature stable at the air conditioner set temperature while reducing energy consumption;

[0043] S3: Obtain the indoor temperature data at the current second monitoring time point. The time interval between every two second monitoring time points is much shorter than the time interval of the control cycle, enabling high-frequency collection of indoor temperature data, thereby achieving rapid adjustment of the wind speed. Determine whether the two-way valve is in the open state at the current time. If the two-way valve is in the open state, compare the indoor temperature data at the currently obtained second monitoring time point with the predicted target temperature T. If the indoor temperature data is greater than T + 0.5, adjust the wind speed to high wind speed. At high wind speed, the refrigeration effect of the fan coil unit will be better, so that when the indoor temperature is high, it can quickly drop to the air-conditioning set temperature. If the indoor temperature data is greater than T and less than T + 0.5, adjust the wind speed to medium wind speed. Since the temperature is already relatively close to the air-conditioning set temperature at this time, too high a wind speed will cause the indoor temperature to drop too quickly and increase energy consumption at the same time. Therefore, only medium wind speed is needed, which not only ensures the refrigeration effect of the air conditioner but also enables the indoor temperature to approach the air-conditioning set temperature and saves energy. If the current indoor temperature is less than T, adjust the wind speed to low wind speed. Since the indoor temperature is already lower than the air-conditioning set temperature at this time, if the temperature continues to drop rapidly, it may cause the temperature to be too much lower than the air-conditioning set temperature, affecting the experience of indoor personnel. Therefore, only low wind speed is needed to prevent the temperature from dropping too quickly.

[0044] If the two-way valve is in the closed state, adjust the wind speed to low wind speed. Since the two-way valve is closed, there is no refrigeration source at this time. Under different wind speeds, the temperature rise speed is about the same. Therefore, only low wind speed is needed to save energy.

[0045] The target temperature prediction model is established based on the linear regression model. The linear regression model is an analysis method in statistics. It predicts or estimates the value of the dependent variable by establishing a linear relationship between one or more independent variables and the dependent variable. This model assumes a linear dependence relationship between variables. The target temperature prediction model uses a formula to predict the value y of the predicted target temperature. x1, x2, x3, and x4 are all input features, namely indoor temperature data, outdoor temperature data, indoor humidity data, and outdoor humidity data. β0 is the intercept term. The core role of the intercept term is to determine the expected value of the dependent variable when all independent variables are zero, thereby providing a reference point for the model so that the model can more accurately reflect the relationship between variables and can also make effective predictions even when the independent variables are zero. β1, β 2、 β3, and β4 are the weights of each input feature respectively.

[0046] Regarding the training of the target temperature prediction model, the specific operation steps are as follows:

[0047] Obtain a number of model training samples. Each model training sample includes indoor temperature data, outdoor temperature data, indoor humidity data, outdoor humidity data, and the air conditioner set temperature at the corresponding moment. Divide all model training samples into a model training set and a model test set. Use the model training set to train the target temperature prediction model with the goal of minimization, and estimate the model parameters, including the intercept term and the weight coefficients of each independent variable, where n is the number of the model training set; i = 1, 2,..., n. Use the model test set to evaluate the target temperature prediction model and obtain the evaluation result. Set the training condition. If the evaluation result meets the training condition, output the trained target temperature prediction model; otherwise, continue to train using the model training set.

[0048] Regarding the setting of the control period, the specific operation steps are as follows:

[0049] Set the update period. Since the indoor heating or cooling speed is different under different environmental data, it is necessary to update the duration of the control period in each update period. The update period can be set to half a day. At the beginning of each update period, record the time required for the indoor temperature to rise from T - 1 to T + 1 when the two-way valve is closed and the low wind speed is on, and use the obtained time as the upper limit duration of the control period. The purpose here is to prevent the indoor temperature from exceeding the air conditioner set temperature too much when the opening time of the two-way valve is very short or always closed during a control period, which affects the temperature control accuracy. With the goal of maximizing the temperature control accuracy and minimizing the on-off frequency of the two-way valve, use the swarm optimization algorithm to search for the optimal control period within the range of 0 to the upper limit duration, and use the obtained optimal control period as the control period of the current update period. Maximizing the temperature control accuracy is to ensure that the indoor temperature is as close as possible to the target temperature; minimizing the on-off frequency of the two-way valve is to reduce energy consumption and extend the equipment life, and prevent the on-off frequency of the two-way valve from being too high.

[0050] In the operation of setting the control period, the swarm optimization algorithm used is the particle swarm optimization algorithm. The particle swarm optimization algorithm is a swarm intelligence optimization technology that simulates the foraging behavior of bird flocks. By randomly initializing a group of particles in the search space and dynamically updating the particle positions and velocities according to individual experience and group experience, iteratively search for the optimal solution. The specific operation steps are as follows:

[0051] A1: Randomly generate a particle swarm within the range of 0 to the upper limit duration. The position of each particle represents a candidate control period, and the position of the particle represents a potential solution of the control period.

[0052] A2: For any particle, initialize the velocity of the particle, which represents how fast the particle moves in the search space, and take the current position as the individual best position of the particle, that is, the best control period found by the particle so far; calculate the fitness value F of the particle using the formula, where D represents the difference between the maximum temperature and the minimum temperature in the current candidate period, and L represents the duration of the current candidate period; and are the weights of and L respectively;

[0053] A3: Traverse all particles, and take the position corresponding to the particle with the highest fitness value F as the global best position, that is, the best control period found among all current particles;

[0054] A4: For any particle, update the velocity and position of the particle based on the dynamic adjustment of individual experience and group experience, and recalculate the fitness value; if the updated fitness value is higher than the fitness value of the individual best position of the particle, then update the individual best position of the particle to the current position;

[0055] A5: Traverse all particles, if there exists any particle whose fitness value of the individual best position is higher than the fitness value of the global best position, then update the global best position to the individual best position of the particle;

[0056] A6: Repeat steps A4 - A5 until the preset maximum number of iterations is reached, and the obtained global best position is the optimal control period.

[0057] The specific operation of calculating and obtaining the output P using the PID control algorithm in step S2:

[0058] Calculate the difference between the current indoor temperature data and the predicted target temperature, and calculate and obtain the PID output P using the formula; where K p is the proportional gain, and the proportional gain is the proportional coefficient for adjusting the output according to the amplitude of the difference; T i is the integral time constant, which is used to calculate the cumulative effect of the deviation over time and helps eliminate the steady-state error of the system; T d is the derivative time constant, which is used to predict the rate of change of the deviation so as to quickly respond to the changes of the system; m represents that there are m control periods within the integral time constant, j = 1, 2,..., m; represents the difference between the indoor temperature data obtained in the jth control period and the predicted target temperature in the most recent time with a duration of the integral time constant; is the difference between the indoor temperature data obtained in the previous control period and the predicted target temperature, and this value is used to calculate the derivative term to help predict the change trend of the deviation; is the duration of the control period.

[0059] Convert the output P into the duty cycle of the two-way valve opening time; the specific operation of applying the duty cycle of the two-way valve opening time within the current preset control period:

[0060] Obtain the range of the PID output P [P min , P max , and use the formula to calculate the duty cycle N of the on-off time of the two-way valve. The duty cycle refers to the ratio of the on-state time of the two-way valve to the total time within the control period and is a key parameter for controlling the operating state of the two-way valve; within the current control period, after the two-way valve is opened, it is then closed; the above operations are repeated in each control period to ensure that the on-off state of the two-way valve can be dynamically adjusted according to the real-time indoor temperature and the output of the PID controller.

[0061] Proportional gain K p , integral time constant T i and derivative time constant T d are all adjusted by the particle swarm optimization algorithm; in order to achieve control objectives such as fast response, stable control, and minimum overshoot, it is necessary to adjust the proportional gain K p , integral time constant T i and derivative time constant T d ; when facing a complex and changing system, the traditional method of manually adjusting parameters may be difficult to achieve the ideal control effect; therefore, through an automated optimization algorithm, more suitable parameter settings can be found more efficiently.

[0062] It should be understood that those of ordinary skill in the art can make improvements or transformations according to the above description, and all such improvements and transformations shall fall within the protection scope of the appended claims of the present invention. The parts not described in detail in this specification belong to the prior art well-known to those of ordinary skill in the art.

Claims

1. Intelligent and comfortable energy-saving fan coil control system, characterized in that, Including: A target temperature prediction module, which is used to input a target temperature prediction set composed of indoor temperature data, outdoor temperature data, indoor humidity data, and outdoor humidity data into a target temperature prediction model, and output a predicted target temperature T. The predicted target temperature T is the air conditioner set temperature within the currently preset control period; A two-way valve control module, which is used to calculate and obtain an output P by using a PID control algorithm according to the difference between the current indoor temperature data and the predicted target temperature; convert the output P into the duty cycle of the two-way valve switching time; within the currently preset control period, apply the duty cycle of the two-way valve switching time to control the opening and closing time of the two-way valve; when the two-way valve is closed, the system has no refrigeration source; A wind speed adjustment module, which is used to adjust the wind speed according to the indoor temperature data and the operating state of the two-way valve; when the two-way valve is in the open state, compare the indoor temperature data at the currently obtained second monitoring time point with the predicted target temperature T. If the indoor temperature data is greater than T + 0.5, adjust the wind speed to high wind speed; if the indoor temperature data is greater than T and less than T + 0.5, adjust the wind speed to medium wind speed; if the current indoor temperature is less than T, adjust the wind speed to low wind speed; when the two-way valve is in the closed state, adjust the wind speed to low wind speed; An intelligent and comfortable energy-saving fan coil control method applied to the intelligent and comfortable energy-saving fan coil control system, and the specific steps are as follows: S1: Obtain the indoor temperature data, outdoor temperature data, indoor humidity data, and outdoor humidity data at the current first monitoring time point, and form a target temperature prediction set with the obtained indoor temperature data, outdoor temperature data, indoor humidity data, and outdoor humidity data; input the target temperature prediction set into the target temperature prediction model, and output a predicted target temperature T. The predicted target temperature T is the air conditioner set temperature within the currently preset control period; S2: Based on the indoor temperature data obtained at the current first monitoring time point, calculate the difference between the current indoor temperature data and the predicted target temperature, and calculate and obtain an output P by using a PID control algorithm; Convert the output P into the duty cycle of the two-way valve switching time; within the currently preset control period, apply the duty cycle of the two-way valve switching time to control the opening and closing time of the two-way valve; S3: Obtain the indoor temperature data at the current second monitoring time point, and judge whether the two-way valve is in the open state at the current time; if the two-way valve is in the open state, compare the indoor temperature data at the currently obtained second monitoring time point with the predicted target temperature T. If the indoor temperature data is greater than T + 0.5, adjust the wind speed to high wind speed; if the indoor temperature data is greater than T and less than T + 0.5, adjust the wind speed to medium wind speed; if the current indoor temperature is less than T, adjust the wind speed to low wind speed; If the two-way valve is in the closed state, adjust the wind speed to low wind speed; Regarding the setting of the control period, the specific operation steps: Set the update period. At the beginning of each update period, record the time required for the indoor temperature to rise from T - 1 to T + 1 when the two-way valve is closed and the low wind speed is on. Take the obtained time as the upper limit duration of the control period. Aiming at maximizing the temperature control accuracy and minimizing the on-off frequency of the two-way valve, use the swarm optimization algorithm to search for the optimal control period within the range of 0 to the upper limit duration, and take the obtained optimal control period as the control period of the current update period.

2. The intelligent and comfortable energy-saving fan coil control system according to claim 1, wherein, The target temperature prediction model is established based on the linear regression model. The target temperature prediction model uses the formula to predict the value y of the predicted target temperature; x1, x2, x3, and x4 are all input features, namely indoor temperature data, outdoor temperature data, indoor humidity data, and outdoor humidity data; β0 is the intercept term; β1, β2, β3, and β4 are the weights of each input feature respectively.

3. The intelligent and comfortable energy-saving fan coil control system according to claim 2, wherein Regarding the training of the target temperature prediction model, the specific operation steps are as follows: Obtain a number of model training samples. Each model training sample includes indoor temperature data, outdoor temperature data, indoor humidity data, outdoor humidity data, and the air conditioner set temperature at the corresponding moment; divide all model training samples into a model training set and a model test set; use the model training set to train the target temperature prediction model with the goal of minimization; where n is the number of the model training set; i = 1, 2,..., n; Use the model test set to evaluate the target temperature prediction model and obtain the evaluation result; set the training conditions. If the evaluation result meets the training conditions, output the trained target temperature prediction model; otherwise, continue to train using the model training set.

4. The intelligent and comfortable energy-saving fan coil control system according to claim 3, characterized in that, In the operation of setting the control period, the swarm optimization algorithm used is the particle swarm optimization algorithm. The specific operation steps are as follows: A1: Randomly generate a particle swarm within the range of 0 to the upper limit duration. The position of each particle represents a candidate control period. A2: For any particle, initialize the velocity of the particle and take the current position as the individual best position of the particle; calculate the fitness value F of the particle using the formula, where D represents the difference between the maximum temperature and the minimum temperature within the current candidate period, and L represents the duration of the current candidate period; and are the weights of and L respectively. A3: Traverse all particles and take the position corresponding to the particle with the highest fitness value F as the global best position. A4: For any particle, update the velocity and position of the particle and recalculate the fitness value; if the updated fitness value is higher than the fitness value of the individual best position of the particle, update the individual best position of the particle to the current position. A5: Traverse all particles. If there exists any particle whose fitness value of the individual best position is higher than the fitness value of the global best position, update the global best position to the individual best position of the particle. A6: Repeat steps A4 - A5 until the maximum number of iterations is reached. The obtained global best position is the optimal control period.

5. The intelligent and comfortable energy-saving fan coil control system according to claim 4, wherein, The specific operation of calculating and obtaining the output P using the PID control algorithm in step S2: Calculate the difference between the current indoor temperature data and the predicted target temperature, and use the formula to calculate the PID output P; where K p is the proportional gain; T i is the integral time constant; T d is the derivative time constant; m represents that there are m control cycles within the integral time constant, j = 1, 2,..., m; represents the difference between the indoor temperature data obtained in the jth control cycle and the predicted target temperature in the most recent time with a duration of the integral time constant; is the difference between the indoor temperature data obtained in the previous control cycle and the predicted target temperature; is the duration of the control cycle.

6. The intelligent and comfortable energy-saving fan coil control system according to claim 5, wherein Convert the output P to the duty cycle of the two-way valve switching time; within the currently preset control cycle, the specific operation of applying the duty cycle of the two-way valve switching time: Obtain the range of the PID output P [P min , P max , and calculate the duty cycle N of the two-way valve opening and closing time using the formula; within the current control cycle, after the two-way valve is opened, it is then closed.

7. The intelligent and comfortable energy-saving fan coil control system according to claim 6, characterized in that, Proportional gain K p , integral time constant T i and derivative time constant T d The values of are all obtained by the particle swarm optimization algorithm.

Images